Quality Management - Statistical Process Control
Quality Management - Statistical Process Control
Quality Management
• Quality
• The ability of a product or service to consistently meet
or exceed customer expectations
• For a decade or so, quality was an important focal point in business. After a
while, this emphasis began to fade as other concerns took precedence
• There has been a recent resurgence in attention to quality given recent
experiences with the costs and adverse attention associated with highly
visible quality failures:
• Auto recalls
• Toys
• Pharmaceuticals
Definitions of Quality
W.
Joseph Philip
Edwards
Juran Crosby
Deming
(1974) (1979)
(1950) In your eyes i.e. Conformance to
Fitness for use
customer requirements
Focus on
Costs of quality Quality is Free
process quality
Juran’s trilogy-
planning,
PDCA Cycle Zero Defects
control &
improvement
Lowe et al. (1986) - Three preachers, One religion
Quality Contributors
• Walter Shewart
• “father of statistical quality control”
• Control charts
• Variance reduction
• Genichi Taguchi
• Taguchi loss function
Assurance – knowledge
Time – the speed with exhibited by personnel Courtesy – the way
which the service is and their ability to customers are treated
delivered convey trust and by employees
confidence
• Plan
Study Do
• Begin by studying and documenting the current process.
• Collect data on the process or problem
• Analyze the data and develop a plan for improvement
• Specify measures for evaluating the plan
• Do
• Implement the plan, document any changes made, collect data for
analysis
PDSA Cycle
• Plan-Do-Study-Act (PDSA) Cycle Act Plan
• Study
• Evaluate the data collection during the do phase Study Do
Productivity
Absenteeism
(c) Cause-and-Effect Diagram: A tool that identifies process
elements (causes) that might effect an outcome
Cause
Materials Methods
Effect
Manpower Machinery
(d) Pareto Chart: A graph to identify and plot problems or defects in
descending order of frequency
Frequency
Percent
A B C D E
(e) Flowchart (Process Diagram): A chart that describes the steps in a
process
(f) Histogram: A distribution showing the frequency of occurrences of a
variable
Distribution
Frequency
Target value
Time
Statistical Process Control
Variations
Variables Attributes
◼ Measured values; e.g., weight, ◼ Has or Has not/Good or Bad/Pass
length, volume, voltage, current or Fail/Accept or Reject
etc. ◼ Characteristics for which you focus
◼ May be in whole or in fractional on defects
numbers ◼ Categorical or discrete random
◼ Continuous random variables variables
Statistical Process Control
x i
where x = i =1
n
xi = observation of a quality characteristic (such as time)
n = total number of observations
x = mean
Sampling Distributions
where = (x − x)
2
x
2
i −
n
i
i
or =
n −1 n −1
σ = standard deviation of a sample
Sample and Process Distributions
Mean
Distribution of
sample means
Process
distribution
25 Time
Relationship Between the Distribution of Sample Means and the Process Distribution
Assignable Causes
Average
Time
(a) Location
Effects of Assignable Causes on the Process Distribution for the Lab Analysis Process
Assignable Causes
Average
Mean
X
Variables Attributes
Charts Charts
R X P C
Chart Chart Chart Chart
Process is out of control ..When ?
Control Charts - Errors
1. . Collect data
4. Plot the sample ranges. If all are in control, proceed to step 5. Otherwise,
find the assignable causes, correct them, and return to step 1.
5. Calculate x for each sample
Steps for x- and R-Charts
7. Plot the sample means. If all are in control, the process is in statistical
control.
8. Continue to take samples and monitor the process. If any are out of
control, find the assignable causes, correct them, and return to step 1. If
no assignable causes are found, assume out-of-control points represent
common causes of variation and continue to monitor the process.
An Alternate Form
σx = σ/ n
σ = standard deviation of the process distribution
n = sample size
x = central line of the chart
z = normal deviate number
Example- Mean chart
How Mean and Range charts complement each other
How mean and Range charts complement each other
Selection of Rational Samples for X & R Chart
• Sample
• From homogeneous lots: Drawn from same population
• Minimize the difference within samples
• Maximize the differences between samples: Maximize the opportunity for variation
from one sample to another.
• Sample should be taken at equal intervals
• Sample Size
• Between 4 to 10, 4 to 5 sample size is common in industry
• Larger the samples size, the better the chance of detecting small shift: More sensitive to
small variation in the process average.
• Frequency of Sampling
• Trade off between the cost of obtaining information compared to the cost of not
detecting a nonconformance items
Process Capability
Process distribution
Lower Upper
specification specification
Minutes
20 25 30
Process distribution
Lower Upper
specification specification
Minutes
20 25 30
Six sigma
Four sigma
Two sigma
Lower Upper
specification specification
Mean
Process Capability
• The process capability index measures how well a
process is centered and whether the variability (if
any) is acceptable
X X
Process
on target with
Reduce low variability Center
spread process
XX
XX
X
X XX
Determining Process Capability
Step 1. Collect data on the process output, and
calculate the mean and the standard deviation
of the process output distribution.
26.2 – 20.0
Lower specification calculation = = 1.53
3(1.35)
The process variability did not meet the four-sigma target of 1.33.
Consequently, she initiated a study to see where variability was
introduced into the process.
Two activities, report preparation and specimen slide preparation,
were identified as having inconsistent procedures. These
procedures were modified to provide consistent performance. New
data were collected and the average turnaround was now 26.1
minutes with a standard deviation of 1.20 minutes.
Assessing Process Capability
She now had the process variability at the four-sigma level of
performance, as indicated by the process capability ratio:
30.0 – 20.0
Cp = = 1.39
6(1.20)
X X
Process
on target with
Reduce low variability Center
spread process
XX
XX
X
X XX